Bring in the Clowns - The Effect of Humor on Health Care
How to cite
Grimm, Christine; Gräser, Melanie; Hoffmann, Roman.
2022. "Bring in the Clowns - The Effect of Humor on Health Care." AEA RCT Registry. AEARCTR-0009770.
https://doi.org/10.1257/rct.9770-2.0
Download:
BibTeX
·
RIS— for reference managers without a browser
translator. Zotero and most
others pick up this page's citation metadata directly.
Christine Grimm*
, Melanie Gräser
, Roman Hoffmann
This project evaluates the effect of clown visits in hospitals on the physical and psychological well-being of patients, as assessed by their caregivers, and health care utilization in a middle-income developing country. The authors conduct a field study in cooperation with an international organization for clown visits operating in the Middle East. By randomizing the schedule of standardized interventions in hospitals, short-term hospitalized children are randomly assigned to two groups: clown visit and no clown visit. Field workers approach accompanying caregivers of both groups for consent to participate in the study and their agreement to take part in a telephone interview, held about ten days after recruiting. A comparison of the answers given in the interviews and the accompanying hospital data will shed light on the immediate effects of clown visits and potential spill-over effects on trust in the healthcare system.
Intervention
Standardized clown bed-site visits in hospitals as established in the field.
Experimental design
Clowns visit hospitals on a randomized schedule. At one (two) out of seven days, there will be a clown visit. At six (five) out of seven days, there will be no clown visit. Naturally, some children will see a clown and others do not. Data collectors independently approach caregivers of both groups for consent to participate in the study. Participants are contacted for a telephone interview and data on the patient is collected from the hospital.
Design details
Randomization method
The randomization was done with the computer using stata.
The randomization can be replicated with the "generate random_number" command for the hospitals in each week. The random number is assigned to seven groups of equal size representing the weekdays. The starting seed is 1234. For every week an additional 10 is added to the seed.
Hospitals that entered our sample group after the start date (including those with two visits) follow an individual do-file that is set up in the same way, using a different starting seed.
Randomization unit
Day of clown visit per week and hospital.
Clusters
We plan to have a threefold analysis. A) treats every patient/caregiver as one observation, B) controls for the day of arrival and hospitals to account for spill-over effects, C) treats the daily situation at each hospital as one observation.
A) 2000 patients / caregivers
B) 375 days of arrival and hospital (5 hospitals x 15 days/month(rough estimate) x 5 months) (in case of adjustments: additional hospitals, fewer days per month covered)
C) 5 hospitals ( in case of adjustments: additional hospitals)
Observations
A) 2000 patients / caregivers
B) 2000 patients / caregivers
C) 500 days (5 hospitals x 20 days/month(rough estimate) x 5 months)
Arms
A)1000 patients / caregivers control + 1000 patients / caregivers treatment
B)1000 patients / caregivers control + 1000 patients / caregivers treatment
C) 428 days and hospital control + 71 days and hospital treatment
Primary outcomes
Our outcome variables are
A) answers in the telephone interview with caregivers and
B) hospital data
to learn about
a) recovery of the patient
b) perception of the hospital stay and
c) perception of the public health care system
Update June 2023:
Despite initial approval from the Ministry of Health, project partner and hospital, access to the administrative data from the hospital was ultimately denied due to legal reasons at the end of the data collection. Our analysis has to adapt accordingly.
Cited by
No citation data available for this registration yet.